WO2007014745A1 - Reseau de communication et ses procedes d'acheminement de paquets de donnees et de localisation et securisation des donnees d'une ressource designee - Google Patents

Reseau de communication et ses procedes d'acheminement de paquets de donnees et de localisation et securisation des donnees d'une ressource designee Download PDF

Info

Publication number
WO2007014745A1
WO2007014745A1 PCT/EP2006/007582 EP2006007582W WO2007014745A1 WO 2007014745 A1 WO2007014745 A1 WO 2007014745A1 EP 2006007582 W EP2006007582 W EP 2006007582W WO 2007014745 A1 WO2007014745 A1 WO 2007014745A1
Authority
WO
WIPO (PCT)
Prior art keywords
node
ring
network
nodes
routing
Prior art date
Application number
PCT/EP2006/007582
Other languages
English (en)
Inventor
Giscard Wepiwe
Plamen L. Simeonov
Original Assignee
Technische Universität Berlin
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Technische Universität Berlin filed Critical Technische Universität Berlin
Priority to US11/989,452 priority Critical patent/US7660320B2/en
Publication of WO2007014745A1 publication Critical patent/WO2007014745A1/fr

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L63/00—Network architectures or network communication protocols for network security
    • H04L63/12—Applying verification of the received information
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L12/00—Data switching networks
    • H04L12/28—Data switching networks characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks]
    • H04L12/42—Loop networks
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/02—Topology update or discovery

Definitions

  • a communication network a method of routing data packets in such communication network and a method of locating and securing data of a desired resource in such communication network
  • the invention regards a communication network with a plurality of network nodes, a method of routing data packets in such communication network and a method of locating and securing data of a desired resource in such communication network. More particularly, the invention regards a communication network and methods for efficient and effective distribution and discovery of recources in large-scale dynamic distributed environments, in particular peer-to-peer systems .
  • Peer-to-peer (P2P) systems are dynamic distributed systems in which nodes of equal roles and capabilities exchange information and services directly with each other. Due to its main design principle of being completely decentralized and self organizing - opposed to the Internet's "traditional" Client-Server paradigm - the P2P-concept emerges to a major design pattern for future applications, system components and infrastructural services, particularly with regard to reliability and resilience.
  • the Napster network is an example of P2P system, which emerged very fast as a content sharing infrastructure (mainly for MP3 music) among a geographically distributed user community.
  • Other popular examples are Gnutella and Fast Track.
  • P2P file sharing systems become increasingly- popular, scalability and resilience have been recognized as the central challenge in designing such systems.
  • Napster uses centralized directory service, which represents a single point of failure.
  • Systems based on Gnutella use flooding to avoid the problem of Napster, but flooding is known to be highly bandwidth consuming.
  • deletion of a node or a group of nodes in a large P2P network may cause disconnection of a node or a subset of nodes from the network.
  • maintaining the network always connected may imply an active or reactive exchanging of control information, which may also be bandwidth-consuming .
  • FastTrack proposes an improvement to Gnutella's flooding, by introducing the notion of nodes differentiation.
  • FastTrack defines some nodes as "supernodes” , which are mostly fast computers with large computing resources, acting as temporally indexing server for other eventual slower nodes. Discovery messages are routed only through supernodes who are more susceptible to know the location of a requesting resource. FastTrack uses a kind of
  • a series of P2P protocols such named Chord, Tapestry, Pastry, CAN, Viceroy, ODRI, CAN-D2B, distance halving, Broose, Koorde and Pagoda have been deployed recently in the research community, which are based on application layer overlay networking construction, to address the problem of routing and lookup efficiency as well as the resilience issue during network churns.
  • the minimum value of the resource lookup distance in such networks generally varies between log d N and log N, N denoting the total number of nodes in a network, and d denoting the degree of a network, the degree being defined as the maximal number of outgoing links that is maintained by any node in the network.
  • Chord A scalable peer-to-peer lookup service for internet applications, SIGCOMM Comput . Commun. Rev., 31 (4) : 149-160, 2001, describe a peer-to-peer network in which a key identifying a resource or desired data item is mapped onto a network node.
  • N. G. de Bruijn A combinatorial problem, In Nederl . Akad . Wetensh. Proc . 49, pages 768-764, 1946, describes now called De Bruijn digraphs which are usually defined as directed graphs that are constructed on an alphabet. The vertices of the graphs are the words of this alphabet. A vertex is formed of D letters chosen from a set with cardinality d.
  • Figure 10 illustrates an example of a B (2, 3) static the De Bruijn network with 8 nodes.
  • each node in the network has exactly a constant number d of incoming and outgoing edges. That is, the routing table size of nodes in the network is constant.
  • D log d N hops between any two nodes in the network.
  • Studies on the connectivity of the De Bruijn networks show that at most d- 1 nodes can fail without disconnecting any node or any set of nodes from the network.
  • the degree d is an important design parameter that can be varied to improve connectivity.
  • the pseudo- function prefix () returns as result the k first elements right from R, with k ⁇ _ D. Routing is achieved by a consecutive left shifting of x with R until a node y is reached.
  • a next -hop node y is chosen after left shifting of x so that y is a neighbor of x.
  • routing in this way in the De Bruijn graph needs at most D hops from source to destination.
  • the message routing from 001 to 110 takes the path 001 -> 011 -> 111 -> 110 as depicted in figure 11.
  • a communication network comprising a plurality of network nodes; wherein the network nodes are logically arranged to form a plurality of concentric rings, all rings except the outermost ring being with high probability fully constructed; - the number of nodes on each fully constructed ring being d D , wherein d denotes the base of an integer having D digits, D ⁇ 1, and wherein D also denotes the ring number, D starting with the value 1 for the innermost ring; each node being identified by an identification number unique in the network, said identification number having D digits, D > 1, and D denoting the ring number on which the node is located, and the value of each digit being between 0 and d-1; and each node of a given fully constructed ring is allowed to route data packets to one node in the next inner ring, to d nodes in the next outer ring, if the next outer ring is fully constructed, and to at least d nodes on the same ring.
  • d can also be
  • HiPeer a highly reliable network of independent peers that will be referred to as HiPeer .
  • the inventive network is a logical network and does not describe the physical arrangement of network nodes.
  • the network nodes will be computers located at user premises and connected via the internet.
  • each ring of the network is logically so constructed that the nodes on a fully constructed ring form a (d, D) De Brujn network. Routing between nodes on such ring is carried out to De Brujn neighbours as is known in the art .
  • the network further comprises computing means initiating a stabilization routine on a finite portion of the network when a node on an inner ring leaves the network, the stabilization routine logically rearranging some of the nodes such that they change their logical ring in a manner that, at the end of the routine, all rings except the outermost ring are fully constructed again.
  • computing means placing a node joining the network on the outermost ring and assigning a unique identification number, and, if the outermost ring is already fully constructed, creating a new outermost ring and placing the joining node on the new outermost ring.
  • the network management costs at a high rate of node "join and leave" in such dynamic distributed network environment is of deterministic magnitude order O(d), with each node needing only at most 2d + 2 finger entries even in churn situation.
  • a node joining the network divides the current largest space between any two nodes in the outermost ring in two spaces of equal size. This ensures an even distribution of keys at all times in the system, the distance between any two neighboring nodes on the rings being as equal as possible .
  • the first node of a new outermost ring is assigned an identification number consisting of D max digits of value zero, D max ⁇ 1, where D max denotes the ring number of the outermost ring.
  • Said first node represents a ring contact node.
  • Said ring contact node of a new outermost ring is present as long as this ring is present.
  • Such ring contact node of a new outermost ring represents a contact point for any node wishing to join the network.
  • a node on a given fully constructed ring has the identification number X ⁇ -iXD- 2 ...XiX 0 , with Xi being a digit between 0 and d-1, and D denoting the number of the ring on which the given node is located.
  • the node is allowed to route data packets to the one node in the next inner ring which has the identicication number x D-1 x D _ 2 ... Xi, i.e.
  • a node on a given fully constructed ring has the identificationiuon number X 0 -XX 0-2 ... XiX 0 , with xi being a digit between 0 and d-1, and D denoting the number of the ring on which the given node is located, the node is allowed to route data packets on the same ring - to all of its De Brujn neighbours,
  • each network node includes a routing table with at most 2d + 2 entries.
  • the invention further regards a method of routing data packets in a communication network as defined in claim 1 between a source node and a destination node.
  • the method discriminates between cases in which the source node and the destination node reside on the same or different rings.
  • the nodes on a fully constructed ring form a (d, D) De Brujn network wherein, if the destination node and the source node reside on the same ring, horizontal routing of the data packets is carried out to De Brujn neighbours.
  • the invention further regards a method of locating and securing data from a desired resource in a communication network as defined in claim 1.
  • the method comprises the steps of assigning a resource at a specific node a global unique identifier (also referred to as GUIDE) , - assigning the resource at least one ambassador which contains the address and the global unique identifier of the specific node, choosing the ambassador with a method which is a function of the global unique identifier of the resource, - calculating at a requesting node the identification number of an ambassador, routing data between the requesting node and the ambassador according to the routing method of claim 12, transferring data between the requesting node and the specific node.
  • each node in the network has at least one and at most D max ambassadors in the network, where N ⁇ d and the total number of rings forming the network is D max .
  • the Global Unique IDEntifier can, for example, be obtained by using the hashing of the resource's name. For the hashing itself, one can use the well-known SHA-I function with additional strategies to avoid collision of GUIDES.
  • GUID global unique identifier
  • one additional ambassador is placed on each further outer ring as the network grows .
  • Figure 2 shows the network of Figure 1 in a situation when nodes leave and join the network
  • HiPeer concentric multi-ring overlay
  • Figure 4 shows the network of Figure 3 in a situation when nodes leave and join the network
  • Figure 5 shows an exemplary routing table and a location table of node with ID 1000 of the network of Figure 1;
  • Figure 6 shows a routing algorithm for routing data packets in a concentric multi-ring overlay (HiPeer) network
  • Figure 8 shows the average number of recources stored per node in a concentric multi-ring overlay (HiPeer) network with about 1200 nodes in a simulation
  • Figure 9 shows the average number of roucing cable entries per node as a function of the network size in a simulation
  • Figure 1 depicts a concentric multi-ring overlay network, which in the following is also referred to as a HiPeer network.
  • overlay network indicates that the network is a logical network. The following description refers to the logical arrangement of the nodes of the network.
  • an exemplary network consists of 4 concentric rings Ci, C 2 , C 3 and C 4 , C 4 being the outermost ring. All rings except the outermost ring are with high probability fully constructed, i.e., fully occupied.
  • Each node x of the network is identified by an identification number (ID) unique in the network.
  • the identification number has D digits, wherein D denotes the ring number on which the node is located.
  • the value of each digit is between 0 and d- 1, wherein d denotes the base of an integer having D digits (and also stands for the degree of the De Brujn network formed by a ring) .
  • d is equal to 2.
  • the number of nodes x on each fully constructed ring is equal to d D .
  • a node x on a given fully constructed ring has the identification number X D -IXD-2 - • .XiXo, with xi being a digit between 0 and d-1, and D denoting the number of the ring on which the given node is located, the node is allowed to route data packets to the one node in the next inner ring
  • node with identification number x D _iX D _ 2 ...X 1 X 0 is allowed to route data to those d nodes in the next outer ring
  • ⁇ is a digit between 0 and d-1.
  • x D- iX D-2 . - .XiXo is equal to 110, the node is allowed to route data packets to nodes in the next outer ring which have the IDs 1100 and 1101.
  • the organization of the network is the same as of the network of Fig. 1.
  • the ring contact node represents a contact point for any node wishing to join the ring.
  • the ring contact node is assigned an identification number consisting of D digits of value zero, D denoting the ring number. Accordingly, in the example of Fig. 1, the ring contact node for ring C 4 has the ID 0000.
  • Figs. 2 and 4 show situations in which nodes join and leave the network.
  • some nodes may change their logical ring. The intention is to reorganize the network such that all rings but the outermost ring are fully constructed.
  • RID ring identifier
  • the network is constructed and maintained progressively as nodes join and leave the system.
  • the gradually network construction obey the following construction rules:
  • Construction rule 1 The rings are so constructed that nodes on a given inner ring can form a. (d, D) De Bruijn network.
  • the logical concentric multi-ring representation is conceived with the idea that the nodes on the inner rings are more reliable than nodes on the outer ones during the gradual network expansion.
  • the concentric multi-ring topology can also allow routing in multiple directions (i.e. multi-dimensional routing: consider a multi -dimensional 3D or 4D or xD representation of the model, which add new surface for trajectories along the routing path; in the sequel, we consider only "flat" (2 dimensional: 2D) multi-ring topologies) , which improves flexibility and reliability in the case of nodes' failure.
  • Construction rule 2 When a node leaves, a stabilization routine is started on a finite portion of the network, whereby some nodes may change their logical ring and thus their status in the concentric multi-ring topology.
  • Figure 2 gives an illustration of the HiPeer's concentric multi-ring overlay networking in the state of a node's joining and leaving the network (2D case) .
  • Construction rule 3 When a new node joins the network and the most outer ring with RID D already has d D nodes, a new ring with RID D + 1 is created.
  • the network can grow as necessary, there is no upper limit for the maximum number of nodes and hence rings in the network.
  • Construction rul e 4 The ring ' s construction begins always wi th the node x ⁇ i n i t,o ⁇ wi th the lowes t iden tifi er on the ring and the node X ⁇ i n i t , D ⁇ i s presen t as long as the ring wi th RID D exi s ts .
  • the rings are gradually filled.
  • the first ring consists of at most d 1 nodes, the second ring contains at most d 2 nodes and so the k th ring contains d k nodes .
  • N max (d D+1 - d)/(d-l) , with d > 1.
  • Construction rule 5 If a ring with RID D exists, then the node with identifier x ⁇ i n i t,D ⁇ always exists on that ring.
  • Node X ⁇ i n i t , D ⁇ plays the role of ring's contact point during the join operation and ensures a balanced distribution of existing nodes on the ring, by performing key space halving as described below.
  • each node has an ID whose length (number of digits) is associated with the ring identifier.
  • each node in the logical overlay network can hold more than one resource.
  • Each resource is assigned a Global Unique IDEntifier (GUIDE) , which can for example be obtained by using the hashing of the resource's name. For the hashing itself, one can use the well-known SHA-I function with additional strategies to avoid collision of GUIDEs.
  • Each node x can place information about its resource R x of length L at any node y on the same ring and on any node y on the next inner ring which is assigned an identification number which is equal to the first D digits of the global unique identifier (GUIDE) of the recource, D denoting the number of the ring on which the other node y is located.
  • the resource placement is extended to all nodes y on the outer rings for which the identification number is equal to the first D digits of the global unique identifier (GUIDE) of the recource, D denoting the respective ring number.
  • such node y is referred as the ambassador of x for resource R x in the network.
  • each node in the network has at least 1 and at most D max ambassadors in the network, where N ⁇ d and the total number of rings forming the network is D max .
  • any node x has at most D max ambassadors in the network.
  • Node x can route to at most d nodes y on the same ring; these are all its De Bruijn neighbors.
  • Each node x in the network maintains at most 2 tables, as is shown in Figure 5 for the node with ID 1000.
  • a resource location table with the list of pairs (GUIDE, Provider ' s IP Address) that contains the list of the resource providers for any resource R of length L wherein the identification number of a recource provider is equal to the first D digits of the global unique identifier (GUIDE) of the recource .
  • an inner ring could be in an incomplete state. However, if the stabilization routine takes an extremely short time T to resume its task, then any inner ring is always fully constructed w.h.p.
  • a special case is when a message is routed between two nodes x and y on the outermost ring of the network; node x forwards the message through vertical out -in routing in one hop to the node on the next inner ring. That node horizontally forwards the message to its neighbors till the node is reached that can vertically forward the message in one hop to the destination node y on the outer ring.
  • routing between any two nodes in the network can be achieved in at most D H ip e er + 1 hops .
  • a node x looking for a resource R y of length L always first sends its resource discovery request horizontally to the node on the same ring the first digits of which are identical to an ambassador's first digits (an ambassador is calculated as stated above) . With high probability, such node knows a provider of R y . Such node uses the "out_forwarding" routine to forward the request until a possible node y (ambassador) is reached that knows the provider of the resource R y .
  • Resource lookup in the network can find an ambassador of the resource provider in at most D H iPeer hops .
  • a new node when it wants to join the network, it should: (i) determine the RID of the actual outermost ring and (ii) choose a node's identifier (ID) on that ring.
  • ID node's identifier
  • the outermost ring of the concentric multi-ring overlay is considered as the network entry point .
  • the newcomer node x will send a "join request" message to the contact node, asking for the attribution of an ID on the actual outermost ring.
  • all nodes X ⁇ mi t,k ⁇ , k from ⁇ l, ..., D ⁇ play the role of a ring contact node, so that the "join request" message is routed to X jinit
  • the operation of determining the actual outermost ring can involve at most D H ip eer nodes to forward the "join request" message .
  • the newcomer node in the outermost ring divides the current largest space between any two nodes in two spaces of equal size.
  • a node x joins the network as indicated in Figure 4, it divides the space between the nodes 110 and 000 into two equidistant spaces and becomes the node with ID 111.
  • the newcomer node After joining the system, and in order to build its own view of the system, the newcomer node should notify its presence to its neighbors, distribute its re-sources, and collect notification messages received from others nodes. After that, the node is connected and it can request and exchange resources with the rest of the network.
  • the HiPeer network overlay allows with each node joining the network a generation of one message to determine the RID of the outermost ring, plus one resource location message to distribute each of its resources, plus at most d + 1 messages to update its routing table. This makes a total constant number of d + 3 messages, which involves at most D nodes for forwarding .
  • node x For the successor selection operation, node x should not need more than one message to choose one of its neighbors to take its place in the system. Subsequently, the leaving node sends its tables to its successor and leaves the network.
  • node x can also send a "delete notification" to all its ambassadors in the network for any of its resources.
  • sending "delete notification" to all the ambassadors can be a costly operation.
  • this can be chosen as an optional alternative with interesting properties in a sparsely populated network.
  • a node x After a node x has communicated its tables to its successor, it leaves the network and the successor node changes its logical ring to become the new node with ID x.
  • the new node x differs from the old one in two points, the IP address and the provided resources.
  • the new node with ID x has the routing and the location tables of the old node x.
  • the successor publishes its resources to ambassadors on the same ring and on the next inner ring and updates the routing entries of its De Bruijn neighbors. This operation implies a network maintenance cost of mostly d + 2 messages for each node leaving the network.
  • This section is to demonstrate the qualities of the HiPeer network and protocol on the basis of simulation results. The performance evaluation is given for standard conditions in a network with minimal packet loss and less congestion.
  • the network consists of good nodes (normal nodes, participating in the network to download and upload files) and malicious nodes (adversarial nodes, participating in the network to undermine its performance) . It were analyzed different behaviors of a malicious peer in a network using threat models. It is further assumed that any- normal node joining the network first builds its routing and location table and then chooses ambassadors for its resources in the network.
  • the network During the first simulation one joins the network each five seconds until the number N of nodes in the network is 1192, which means that the network is fully constructed with 6 rings.
  • N of nodes in the network is 1192, which means that the network is fully constructed with 6 rings.
  • nodes in the network regularly request existing resources in the network. It is simulated the lookup path length, which is the number of peers that a request message visits before reaching a valid resource provider or a valid ambassador of the requested resource. It is also simulated the routing table size per node , when the number of nodes in the network grows exponentially.
  • the other simulation evaluates the load balancing of the protocol by simulating the number of resources per nodes.
  • Lookup Path Length The principle is to find a provider or an ambassador of a resource in the network.
  • the nodes regularly request an existing resource in the growing network by sending lookup request messages. It was found that there is no lookup failure in the network as long as there is no node failure in the network.
  • Figure 7b it is shown a comparison result of the average path length with the path length in the known Chord protocol; this simulation result shows that resources in a
  • HiPeer based network can be reached within a shortest path length than e.g. in Chord.
  • Load Balancing An important assessment factor that shows the strength of the HiPeer protocol is the load balancing of the system. Following the P2P philosophy, it is important that the nodes in the network have almost the same responsibility of maintaining the network. It is indicated in Figure 8 that available resources are fairly distributed between nodes in the network. This is, when a new node joins the network with some new resources, they are distributed so that no node is overwhelmed when an other one has less ambassador duty. The Figure 8 shows that the average distribution of resources per node is almost constant at any size of the network. This is possible because new nodes joining the network distribute its resources only to the virtually local nodes on the same ring and on the next inner ring.
  • Routing Table Size Generally, ensuring a good connectivity of the nodes and hence of resources in the network implies that nodes are aware of a great amount of nodes in the necwork. So, as the network grows, the routing table in most known protocols logarithmically grows with the number of nodes in the network. In the present HiPeer network, the routing table is limited to 2d + 2 independently of the number of nodes in the network.
  • Figure 9 presents the evolution of the mean routing size per node in a network growing from 0 to 1092 (where 1092 is the number of nodes in a fully constructed network with 6 concentric rings) .
  • the invention is not restricted in its configuration to the exemplary embodiments presented above, which are to be understood as only given by way of example. A person skilled in the art recognizes the existence of numerous alternative variants for the embodiment, which, in spite of their departure from the exemplary embodiments described, make use of the teaching defined in the claims.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Computer Security & Cryptography (AREA)
  • Computer Hardware Design (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)
  • Computer And Data Communications (AREA)

Abstract

La présente invention concerne un réseau de communication constitué d'une pluralité de noeuds logiquement disposés en anneaux concentriques qui tous à l'exception du plus extérieur présente une probabilité d'être complètement construits. Le nombre de noeuds de chaque anneau complètement construit est de dD, d correspondant à la base d'un entier à D chiffres tels que D = 1, D représentant également le numéro de l'anneau en commençant à 1 pour l'anneau le plus intérieur. Chaque noeud est identifié par un numéro d'identification unique dans le réseau. Ce numéro d'identification comporte D chiffres tels que D = 1, D correspondant au numéro d'anneau sur lequel est situé le noeud, la valeur de chaque noeud se situant entre 0 et d-1. Chaque noeud d'un anneau complètement construit est admis à acheminer des paquets de données à destination d'un noeud de l'anneau intérieur voisin, à d noeuds dans l'anneau extérieur voisin si cet anneau extérieur est complètement construit, et à d noeuds du même anneau. L'invention concerne également un procédé de localisation et de sécurisation de données d'une ressource désignée dans un tel réseau de communication, et un procédé permettant de localiser et de sécuriser des données d'une ressource désignée dans un tel réseau de communication.
PCT/EP2006/007582 2005-07-28 2006-07-26 Reseau de communication et ses procedes d'acheminement de paquets de donnees et de localisation et securisation des donnees d'une ressource designee WO2007014745A1 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US11/989,452 US7660320B2 (en) 2005-07-28 2006-07-26 Communication network, a method of routing data packets in such communication network and a method of locating and securing data of a desired resource in such communication network

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
EP05090225A EP1748604B1 (fr) 2005-07-28 2005-07-28 Réseau de communication, procédé de routage des paquets et procédé de localisation et sécurisation de données d'une ressource desirée dans un tel réseau
EP05090225.3 2005-07-28

Publications (1)

Publication Number Publication Date
WO2007014745A1 true WO2007014745A1 (fr) 2007-02-08

Family

ID=36619286

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/EP2006/007582 WO2007014745A1 (fr) 2005-07-28 2006-07-26 Reseau de communication et ses procedes d'acheminement de paquets de donnees et de localisation et securisation des donnees d'une ressource designee

Country Status (5)

Country Link
US (1) US7660320B2 (fr)
EP (1) EP1748604B1 (fr)
AT (1) ATE460032T1 (fr)
DE (1) DE602005019718D1 (fr)
WO (1) WO2007014745A1 (fr)

Families Citing this family (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7961711B2 (en) * 2007-08-06 2011-06-14 Microsoft Corporation Fitness based routing
JP5684110B2 (ja) 2008-05-23 2015-03-11 テレフオンアクチーボラゲット エル エム エリクソン（パブル） ルーティングテーブルを維持する方法およびオーバレイネットワーク内で使用するためのノード
US8996726B2 (en) * 2008-06-19 2015-03-31 Qualcomm Incorporated Methods and apparatus for event distribution and routing in peer-to-peer overlay networks
WO2010014856A2 (fr) * 2008-07-31 2010-02-04 Tekelec Procédés, systèmes et supports lisibles par ordinateur pour étrangler un trafic vers un serveur de réseau de protocole internet (ip) à l'aide d'identifiants de nom d'hôte alias attribués au serveur de réseau ip avec un système de noms de domaine (dns)
KR101409384B1 (ko) * 2010-06-15 2014-06-18 테켈렉, 인코퍼레이티드 다이어미터 네트워크에서 동적 시발 기반 라우팅 키 등록을 제공하기 위한 방법, 시스템 및 컴퓨터 판독 가능한 매체
US8547908B2 (en) 2011-03-03 2013-10-01 Tekelec, Inc. Methods, systems, and computer readable media for enriching a diameter signaling message
US9712559B2 (en) * 2012-05-31 2017-07-18 Red Hat, Inc. Identifying frames
TWI489889B (zh) * 2012-12-28 2015-06-21 Ind Tech Res Inst 內容遞送網路及同儕網路之流量控制方法及系統
RU2526755C1 (ru) * 2013-04-08 2014-08-27 Открытое акционерное общество "Калужский научно-исследовательский институт телемеханических устройств" Способ многомерной динамической маршрутизации в сети связи с пакетной передачей сообщений
US8949762B1 (en) 2013-10-22 2015-02-03 International Business Machines Corporation Computer-based modeling of integrated circuit congestion and wire distribution for products and services
TW201545510A (zh) * 2014-05-30 2015-12-01 Ibm 在分散式計算系統中進行訊息路由的方法
RU2608678C1 (ru) * 2015-11-17 2017-01-23 Федеральное государственное бюджетное образовательное учреждение высшего профессионального образования "Казанский национальный исследовательский технический университет им. А.Н. Туполева - КАИ" (КНИТУ-КАИ) Способ многомерной динамической маршрутизации в сети связи с пакетной передачей сообщений
KR20200138755A (ko) * 2018-03-29 2020-12-10 에스.지.에이. 이노베이션스 엘티디. 메시지/파일을 공개적 및 비밀리에 교환하고 동적 하위 그룹을 생성하기 위한 시스템, 디바이스 및 방법
JP7074018B2 (ja) * 2018-10-22 2022-05-24 日本電信電話株式会社 分散処理システムおよび分散処理方法
RU2737702C1 (ru) * 2020-03-05 2020-12-02 федеральное государственное казенное военное образовательное учреждение высшего образования "Военная академия связи имени Маршала Советского Союза С.М. Буденного" Министерства обороны Российской Федерации Способ динамической маршрутизации трафика в сети связи
CN116886591B (zh) * 2023-09-06 2023-12-15 芯来智融半导体科技(上海)有限公司 计算机网络系统及路由方法

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
ALAN MISLOVE,PETER DRUSCHEL: "Providing Administrative Control and Autonomy in Structured Peer-to-Peer Overlays", 2004, RICE UNIVERSITY,HOUSTON,TX,USA, XP002389577 *
BHANDARI R: "Optimal physical diversity algorithms and survivable networks", COMPUTERS AND COMMUNICATIONS, 1997. PROCEEDINGS., SECOND IEEE SYMPOSIUM ON ALEXANDRIA, EGYPT 1-3 JULY 1997, LOS ALAMITOS, CA, USA,IEEE COMPUT. SOC, US, 1 July 1997 (1997-07-01), pages 433 - 441, XP010241387, ISBN: 0-8186-7852-6 *
DATTA A ET AL: "On de Bruijn routing in distributed hash tables: there and back again", PEER-TO-PEER COMPUTING, 2004. PROCEEDINGS. PROCEEDINGS. FOURTH INTERNATIONAL CONFERENCE ON ZURICH, SWITZERLAND 25-27 AUG. 2004, PISCATAWAY, NJ, USA,IEEE, 25 August 2004 (2004-08-25), pages 159 - 166, XP010724979, ISBN: 0-7695-2156-8 *
LIU Z ED - INSTITUTE OF ELECTRICAL AND ELECTRONICS ENGINEERS: "Optimal routing in the de Bruijn networks", PROCEEDINGS OF THE INTERNATIONAL CONFERENCE ON DISTRIBUTED COMPUTING SYSTEMS. PARIS, MAY 28 - JUNE 1, 1990, LOS ALAMITOS, IEEE COMP. SOC. PRESS, US, vol. CONF. 10, 28 May 1990 (1990-05-28), pages 537 - 544, XP010019300, ISBN: 0-8186-2048-X *
STOICA I ET AL: "CHORD: A SCALABLE PEER-TO-PEER LOOKUP PROTOCOL FOR INTERNET APPLICATIONS", IEEE / ACM TRANSACTIONS ON NETWORKING, IEEE / ACM, NEW YORK, NY, US, vol. 11, no. 1, February 2003 (2003-02-01), pages 17 - 32, XP001144289, ISSN: 1063-6692 *

Also Published As

Publication number Publication date
EP1748604B1 (fr) 2010-03-03
US20090092144A1 (en) 2009-04-09
US7660320B2 (en) 2010-02-09
DE602005019718D1 (de) 2010-04-15
ATE460032T1 (de) 2010-03-15
EP1748604A1 (fr) 2007-01-31

Similar Documents

Publication Publication Date Title
EP1748604B1 (fr) Réseau de communication, procédé de routage des paquets et procédé de localisation et sécurisation de données d'une ressource desirée dans un tel réseau
EP1650911B1 (fr) Rendezvous von Betriebsmittelanforderungen mit den entsprechenden Betriebsmitteln
US20070230468A1 (en) Method to support mobile devices in a peer-to-peer network
Eberspächer et al. Structured p2p networks in mobile and fixed environments
JP4533923B2 (ja) 階層型ピアツーピアシステムにおける負荷バランシング機能を有するスーパーピア及び該スーパーピアを動作させる方法
Cherbal et al. A survey of DHT solutions in fixed and mobile networks
Galluccio et al. Georoy: A location-aware enhancement to Viceroy peer-to-peer algorithm
Li et al. An efficient superpeer overlay construction and broadcasting scheme based on perfect difference graph
Wepiwe et al. A concentric multi-ring overlay for highly reliable p2p networks
Conti et al. Peer-to-peer computing in mobile ad hoc networks
Guo et al. P4P Pastry: A novel P4P-based Pastry routing algorithm in peer to peer network
Zhang et al. Resilient capacity-aware multicast based on overlay networks
Karamachoski et al. BloHeS Island management protocols
Lu et al. A scalable P2P overlay based on arrangement graph with minimized overhead
JP5287500B2 (ja) ノード装置およびノード装置用プログラム
Kellerer et al. Proactive search routing for mobile peer-to-peer networks: Zone-based p2p
Ktari et al. Exploiting power-law node degree distribution in chord overlays
Ngo From inter-connecting P2P overlays to co-operating P2P systems
Ktari et al. Empowering chord DHT overlays
Brands et al. Taxonomy of p2p applications
Fayçal et al. CAP: a context-aware peer-to-peer system
Popescu et al. On the implementation of a content-addressable network
Gündüz Analysis of Popularity Degradation in P2P Overlay Topology
Zhao Decentralized Object Location and Routing: A New Networking Paradigm
Hsu et al. Message transmission techniques for low traffic P2P services

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application
NENP Non-entry into the national phase

Ref country code: DE

WWE Wipo information: entry into national phase

Ref document number: 11989452

Country of ref document: US

122 Ep: pct application non-entry in european phase

Ref document number: 06762914

Country of ref document: EP

Kind code of ref document: A1